NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part2/10407

Internet Message Format | 1996-08-05 | 2.7 KB

Path: keats.ugrad.cs.ubc.ca!not-for-mail
From: [e-mail] (Kazimir Kylheku)
Newsgroups: comp.lang.c
Subject: Re: Pointers!!??
Date: 17 Mar 1996 09:04:50 -0800
Organization: Computer Science, University of B.C., Vancouver, B.C., Canada
Message-ID: <[e-mail]>
References: <4ifhiu$[e-mail]>
NNTP-Posting-Host: keats.ugrad.cs.ubc.ca

In article <4ifhiu$[e-mail]>,
Andrew Heiz  <[e-mail]> wrote:
 >Hi,
 >
 >I have a newbie question about pointers. Will sombody please explain
 >what the advantages of using pointers are. We learned about them in
 >class and programmed with them but we never learned why we were using
 >them. Why would it be more appropriate to use a pointer rather than a
 >variable name? Is it a space/time consideration?

Nearly everything is a space/time consideration :) but in this case it's a
more of a power/simplicity tradeoff.

One advantage of pointers is that they can be dynamically manufactured at run
time. You call malloc() to get a piece of memory and it gives you a new, unique
pointer. Most non-trivial programs need to be written so that they don't impose
compile-time constraints on the size of data sets---the maximum size of the
data may not even be _known_ accurately beforehand; they need to be able to
dynamically allocate new objects.  Pointers are the method in C for keeping
track of such objects.

Another advantage is that pointers can be modified from referring to one object
to refer to another. For example, if you know that a pointer refers to an array
element, you can increment the pointer to refer to the next array element (if
there is one). Or you can use a pointer to walk through a linked data
structure, such as a tree, in which a parent node uses pointers to refer to
children.

Variable names are just compile-time labels, and are not themselves variable.
They are bound to a single object in their scope, and so can't be used for
these kinds of purposes.

Finally, computers typically use physical pointers at the machine language
level.  On machines that use pointers (which is by far the majority of
computers), the C pointer abstraction maps onto the machine's representation: a
word which contains a memory address. Code that uses pointers is often more
succint and efficient, though this is not a rule by any means.

High-level language programs don't have to use pointers explicitly in order to
benefit from them. Compilers will often use pointers to implement high-level
features which don't involve abstracted pointers. For example, a loop which
goes through an array of integers in Pascal, may still generate optimized
machine code that uses pointers to access the memory.
--